Please note: Linked content is NOT stored on Carnegie Mellon University and we can't guarantee its availability, quality, security or accept any liability.
A Computational Model of Internal Control Testing Plan Selection
Any type of content formally published in an academic journal, usually following a peer-review process.
posted on 01.01.1998by James M. Peters, Jefferson T. Davis
Since 1977, the importance of internal control evaluation (ICE) has increased due to
passage of the Foreign Corrupt Practices Act, the Federal Deposit Insurance Corporation
Improvement Act, and other regulatory initiatives. Traditional audit approaches structure the
description and documentation of control systems, but do not provide systematic, precise
evaluation of accounting information system (AIS) reliability. Although researchers have
developed quantitative evaluation methods that provide precise AIS reliability evaluations,
practitioners have not adopted them because they become intractable when applied in practice.
This research develops an optimal, mathematical model of an ICE task, internal control
testing plan selection, that maintains tractability by solving a part of the overall ICE task and by
making simplifying assumptions based on field research. The model selects an optimal controltesting
plan given a description of an AIS and the auditor's desired type of assurance in an account
balance. The model was validated by comparing its testing plans to both experienced auditors and
a professional benchmark. The results indicate that the model's testing plans test sufficient
controls to provide auditors with their desired assurance but do so by testing fewer controls than
either experienced auditors or the professional benchmark.